mirror of
https://github.com/zebrajr/pytorch.git
synced 2026-01-15 12:15:51 +00:00
a478435505
Fixes #153069 Updated ldexp to calculate $a\cdot2^b$ as $2^{\log_2(a) + b}$. This helps fix the issue where $2^b$ is not in range for a specific data type (such as $2^{-25}$ or $2^{20}$ with torch.half) but $a\cdot2^b$ is (such as $2\cdot2^{-25}$ or $2^{-10}\cdot 2^{20}$). The issue was happening because $2^b$ was being calculated first, which would go out of range and mess up the rest of the calculation. Using $2^{\log_2(a) + b}$, we calculate ${\log_2(a) + b}$ first, thus the exponential will only go out of range if the whole expression is out of range. This also fixes the issue #133265, that was closed since it was a duplicate of the other. @ngimel, @isuruf Pull Request resolved: https://github.com/pytorch/pytorch/pull/168357 Approved by: https://github.com/ngimel
This folder contains a number of scripts which are used as
part of the PyTorch build process. This directory also doubles
as a Python module hierarchy (thus the __init__.py).
Overview
Modern infrastructure:
- autograd - Code generation for autograd. This includes definitions of all our derivatives.
- jit - Code generation for JIT
- shared - Generic infrastructure that scripts in
tools may find useful.
- module_loader.py - Makes it easier to import arbitrary Python files in a script, without having to add them to the PYTHONPATH first.
Build system pieces:
- setup_helpers - Helper code for searching for third-party dependencies on the user system.
- build_pytorch_libs.py - cross-platform script that builds all of the constituent libraries of PyTorch, but not the PyTorch Python extension itself.
- build_libtorch.py - Script for building libtorch, a standalone C++ library without Python support. This build script is tested in CI.
Developer tools which you might find useful:
- git_add_generated_dirs.sh and git_reset_generated_dirs.sh - Use this to force add generated files to your Git index, so that you can conveniently run diffs on them when working on code-generation. (See also generated_dirs.txt which specifies the list of directories with generated files.)
Important if you want to run on AMD GPU:
- amd_build - HIPify scripts, for transpiling CUDA
into AMD HIP. Right now, PyTorch and Caffe2 share logic for how to
do this transpilation, but have separate entry-points for transpiling
either PyTorch or Caffe2 code.
- build_amd.py - Top-level entry point for HIPifying our codebase.
Tools which are only situationally useful:
- docker - Dockerfile for running (but not developing) PyTorch, using the official conda binary distribution. Context: https://github.com/pytorch/pytorch/issues/1619
- download_mnist.py - Download the MNIST dataset; this is necessary if you want to run the C++ API tests.